INDEX

Explanations

the word "the" or phrases closely related to it

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NameInMap

-0.61

Hozzáférés

-0.54

 voilà

-0.52

ankan

-0.51

ligg

-0.51

cokinetics

-0.49

 διά

-0.48

plotlib

-0.47

lwjgl

-0.46

pushFollow

-0.46

POSITIVE LOGITS

FROM

1.27

 FROM

1.23

 från

1.08

 From

1.07

From

1.03

 from

1.03

from

0.97

desde

0.97

 từ

0.94

getFrom

0.90

Activations Density 0.126%

the word "the" or phrases closely related to it

No Comments

No Known Activations